SPEC® CFP2006 Result

Supermicro (Test Sponsor: Advanced Micro Devices)

Supermicro A+ Server 1022G-NTF,
AMD Opteron 6278

SPECfp®2006 = 45.2

SPECfp_base2006 = 36.7

CPU2006 license:	49	Test date:	Mar-2012
Test sponsor:	Advanced Micro Devices	Hardware Availability:	Jun-2012
Tested by:	Advanced Micro Devices	Software Availability:	Dec-2011

Hardware
CPU Name:	AMD Opteron 6278
CPU Characteristics:	AMD Turbo CORE technology up to 3.30 GHz
CPU MHz:	2400
FPU:	Integrated
CPU(s) enabled:	32 cores, 2 chips, 16 cores/chip
CPU(s) orderable:	1,2 chips
Primary Cache:	512 KB I on chip per chip, 64 KB I shared / 2 cores; 16 KB D on chip per core
Secondary Cache:	16 MB I+D on chip per chip, 2 MB shared / 2 cores
L3 Cache:	16 MB I+D on chip per chip, 8 MB shared / 8 cores
Other Cache:	None
Memory:	64 GB (8 x 8 GB 2Rx4 PC3-12800R-11, ECC)
Disk Subsystem:	1 x 500 GB SATA, 7200 RPM
Other Hardware:	None

Software
Operating System:	Red Hat Enterprise Linux Server release 6.1, Kernel 2.6.32-131.0.15.el6.x86_64
Compiler:	C/C++/Fortran: Version 4.2.5.2 of x86 Open64 Compiler Suite (from AMD)
Auto Parallel:	Yes
File System:	ext3
System State:	Run level 3 (Full multiuser with network)
Base Pointers:	64-bit
Peak Pointers:	32/64-bit
Other Software:	None

Results Table

Benchmark	Base						Peak
Benchmark	Seconds	Ratio	Seconds	Ratio	Seconds	Ratio	Seconds	Ratio	Seconds	Ratio	Seconds	Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
410.bwaves	63.2	215	65.8	206	63.3	215	39.9	341	39.9	340	40.1	339
416.gamess	1053	18.6	1059	18.5	1052	18.6	990	19.8	997	19.6	992	19.7
433.milc	277	33.1	277	33.1	277	33.1	233	39.4	233	39.5	233	39.4
434.zeusmp	147	61.8	146	62.5	146	62.4	138	66.1	135	67.3	136	67.0
435.gromacs	351	20.4	350	20.4	357	20.0	338	21.1	338	21.1	338	21.1
436.cactusADM	91.2	131	91.0	131	85.1	140	63.3	189	62.6	191	62.8	190
437.leslie3d	365	25.8	367	25.6	367	25.6	356	26.4	356	26.4	355	26.5
444.namd	485	16.5	485	16.5	485	16.5	473	16.9	473	17.0	473	17.0
447.dealII	294	39.0	295	38.8	294	39.0	267	42.9	267	42.8	267	42.9
450.soplex	362	23.0	362	23.0	362	23.0	338	24.6	338	24.7	339	24.6
453.povray	252	21.1	253	21.1	252	21.1	235	22.6	236	22.5	235	22.6
454.calculix	284	29.0	280	29.5	281	29.4	269	30.7	268	30.8	268	30.8
459.GemsFDTD	284	37.3	283	37.5	282	37.7	252	42.1	241	44.0	242	43.8
465.tonto	433	22.7	439	22.4	425	23.2	421	23.4	413	23.8	413	23.8
470.lbm	145	94.5	150	91.8	147	93.2	31.0	443	31.0	443	31.1	442
481.wrf	260	42.9	260	42.9	260	42.9	268	41.7	262	42.7	265	42.1
482.sphinx3	876	22.3	874	22.3	874	22.3	649	30.0	647	30.1	646	30.2

Submit Notes

The config file option 'submit' was used.
'numactl' was used to bind copies to the cores.
See the configuration file for details.

Operating System Notes

'ulimit -s unlimited' was used to set environment stack size
'ulimit -l 2097152'  was used to set environment locked pages in memory limit

Set transparent_hugepage=never as a boot parameter in /boot/grub/menu.lst
Set kernel/randomize_va_space=0 in /etc/sysctl.conf
cpuspeed stop was used to set the CPU frequency to its maximum.

Set vm/nr_hugepages=4000 in /etc/sysctl.conf
mount -t hugetlbfs nodev /mnt/hugepages

General Notes

Environment variables set by runspec before the start of the run:
HUGETLB_LIMIT = "4000"
LD_LIBRARY_PATH = "/root/work/cpu2006v1.2/amd1104-speed-libs-revA/32:/root/work/cpu2006v1.2/amd1104-speed-libs-revA/64"
O64_OMP_AFFINITY_MAP = "0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31"
O64_OMP_SPIN_COUNT = "800000"
O64_OMP_SPIN_USER_LOCK = "true"

The x86 Open64 Compiler Suite is only available from (and supported by) AMD at
http://developer.amd.com/cpu/open64

Binaries were compiled on a system with 2x AMD Opteron 6220 chips + 64GB Memory using RHEL 6.1

Base Compiler Invocation

C benchmarks:

opencc

C++ benchmarks:

openCC

Fortran benchmarks:

openf95

Benchmarks using both Fortran and C:

opencc openf95

Base Portability Flags

410.bwaves:	-DSPEC_CPU_LP64
416.gamess:	-DSPEC_CPU_LP64
433.milc:	-DSPEC_CPU_LP64
434.zeusmp:	-DSPEC_CPU_LP64
435.gromacs:	-DSPEC_CPU_LP64
436.cactusADM:	-DSPEC_CPU_LP64 -fno-second-underscore
437.leslie3d:	-DSPEC_CPU_LP64
444.namd:	-DSPEC_CPU_LP64
447.dealII:	-DSPEC_CPU_LP64
450.soplex:	-DSPEC_CPU_LP64
453.povray:	-DSPEC_CPU_LP64
454.calculix:	-DSPEC_CPU_LP64
459.GemsFDTD:	-DSPEC_CPU_LP64
465.tonto:	-DSPEC_CPU_LP64
470.lbm:	-DSPEC_CPU_LP64
481.wrf:	-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG -fno-second-underscore
482.sphinx3:	-DSPEC_CPU_LP64

Base Optimization Flags

Peak Compiler Invocation

C benchmarks:

opencc

C++ benchmarks:

openCC

Fortran benchmarks:

openf95

Benchmarks using both Fortran and C:

opencc openf95

Peak Portability Flags

410.bwaves:	-DSPEC_CPU_LP64
416.gamess:	-DSPEC_CPU_LP64
433.milc:	-DSPEC_CPU_LP64
434.zeusmp:	-DSPEC_CPU_LP64
435.gromacs:	-DSPEC_CPU_LP64
436.cactusADM:	-DSPEC_CPU_LP64 -fno-second-underscore
437.leslie3d:	-DSPEC_CPU_LP64
444.namd:	-DSPEC_CPU_LP64
447.dealII:	-DSPEC_CPU_LP64
453.povray:	-DSPEC_CPU_LP64
454.calculix:	-DSPEC_CPU_LP64
459.GemsFDTD:	-DSPEC_CPU_LP64
465.tonto:	-DSPEC_CPU_LP64
470.lbm:	-DSPEC_CPU_LP64
481.wrf:	-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG -fno-second-underscore
482.sphinx3:	-DSPEC_CPU_LP64

Peak Optimization Flags

C benchmarks:

433.milc:	-march=bdver1 -Ofast -CG:movnti=1 -CG:locs_best=on -HP:bdt=2m:heap=2m -IPA:plimit=7000 -IPA:callee_limit=1200 -OPT:struct_array_copy=2 -OPT:alias=field_sensitive
470.lbm:	-march=bdver1 -Ofast -mso -apo -CG:sse_cse_regs=0 -LNO:prefetch_ahead=4 -CG:locs_shallow_depth=1 -CG:cmp_peep=on -CG:compute_to=on -OPT:unroll_times_max=8 -OPT:unroll_size=256 -OPT:unroll_level=2 -OPT:keep_ext=on -OPT:alias=restricted -m3dnow -IPA:inline=off
482.sphinx3:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -LNO:loop_model_simd=on -LNO:simd_rm_unity_remainder=on -OPT:malloc_alg=2 -CG:cmp_peep=on -CG:local_sched_alg=2 -CG:use_incdec=off -INLINE:aggressive=on -WOPT:sib=on -HP

C++ benchmarks:

444.namd:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -LNO:ignore_feedback=off -CG:local_sched_alg=2 -CG:load_exe=0 -OPT:unroll_size=256 -fno-exceptions -HP:bdt=2m:heap=2m
447.dealII:	-march=bdver1 -Ofast -LNO:simd=0 -D__OPEN64_FAST_SET -static -INLINE:aggressive=on -OPT:alias=disjoint -OPT:unroll_times_max=8 -OPT:unroll_size=256 -OPT:unroll_level=2 -HP:bdt=2m:heap=2m
450.soplex:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -INLINE:aggressive=on -OPT:RO=1 -OPT:IEEE_arith=3 -OPT:IEEE_NaN_Inf=off -OPT:fold_unsigned_relops=on -fno-exceptions -CG:p2align=0 -m32 -HP:bdt=2m:heap=2m -WOPT:sib=on
453.povray:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -CG:pre_local_sched=off -INLINE:aggressive=on -HP:bdt=2m:heap=2m -OPT:transform=2 -OPT:alias=disjoint -WOPT:aggcm=0

Fortran benchmarks:

410.bwaves:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -apo -OPT:Ofast -OPT:treeheight=on -LNO:blocking=off -LNO:prefetch=2 -LNO:pf2=0 -LNO:prefetch_ahead=3 -LNO:ignore_feedback=off -LNO:fu=4 -LNO:loop_model_simd=on -LNO:simd_rm_unity_remainder=on -WOPT:aggstr=0 -HP:bdt=2m:heap=2m -CG:cmp_peep=on -CG:p2align=0
416.gamess:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -LNO:fu=6 -LNO:blocking=0 -LNO:simd=0 -OPT:Ofast -OPT:ro=3 -OPT:unroll_size=256 -OPT:unroll_times_max=2 -CG:local_sched_alg=1 -HP:bdt=2m:heap=2m -WOPT:sib=on
434.zeusmp:	-march=bdver1 -Ofast -apo -LNO:blocking=off -LNO:interchange=off -LNO:fusion_peeling_limit=0 -OPT:treeheight=on -OPT:unroll_size=256 -CG:cmp_peep=on -CG:compute_to=on -GRA:prioritize_by_density=on -HP:bdt=2m:heap=2m
437.leslie3d:	-march=bdver1 -Ofast -LNO:prefetch=2 -LNO:blocking=off -CG:interior_ptrs=on -OPT:unroll_size=256 -GRA:prioritize_by_density=on -HP:bdt=2m:heap=2m
459.GemsFDTD:	-march=bdver1 -Ofast -OPT:unroll_size=0 -LNO:fission=2 -CG:load_exe=0 -CG:local_sched_alg=2 -HP -apo
465.tonto:	-march=bdver1 -Ofast -OPT:alias=no_f90_pointer_alias -LNO:blocking=off -CG:load_exe=1 -CG:local_sched_alg=1 -IPA:plimit=525 -HP

Benchmarks using both Fortran and C:

435.gromacs:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -OPT:rsqrt=2 -HP:bdt=2m:heap=2m
436.cactusADM:	-march=bdver1 -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -LNO:blocking=off -LNO:prefetch=2 -HP:bdt=2m:heap=2m -CG:locs_shallow_depth=1 -CG:load_exe=0 -WOPT:sib=on -apo
454.calculix:	-march=bdver1 -Ofast -OPT:unroll_size=256 -GRA:optimize_boundary=on -HP:bdt=2m:heap=2m
481.wrf:	-march=bdver1 -Ofast -OPT:unroll_size=256 -LNO:blocking=off -LANG:copyinout=off -IPA:callee_limit=5000 -GRA:prioritize_by_density=on -CG:load_exe=1 -HP -WOPT:sib=on -apo

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/x86-open64-425-flags-speed-revA-I.html,
http://www.spec.org/cpu2006/flags/amd-platform-speed-revA-I.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2006/flags/x86-open64-425-flags-speed-revA-I.xml,
http://www.spec.org/cpu2006/flags/amd-platform-speed-revA-I.xml.